Psychological recovery after disaster – resilience is the norm


This page in:

I’ve been reading a good bit on psychological responses to conflict and disaster for on-going work and am struck by the tone of discussion in the popular press soon after a potentially traumatic event. In these reports, trauma among the survivors is often presumed widespread and the focus is on its expected costs and consequences. However more recent academic work on this topic argues that an exclusive focus on the traumatized misses most of the story.

According to a 2010 review paper by Bonanno and others, this emphasis on trauma morbidity was mirrored in the early academic work on post-traumatic stress disorder (PTSD) which assumed that the near or complete absence of trauma symptoms only occurred in individuals with exceptional emotional strength. In contrast, numerous recent studies have instead indicated that the much more common response to a potentially traumatic event is characterized by what is termed “resilience”.

This resilience is not the complete absence of trauma symptoms – exposure and loss are consistently associated with high levels of reactions to post-traumatic stress such as recurrent intrusive recollections of the traumatizing event. Resilience is instead characterized as adaptability and a “bouncing back” after a harmful event rather than total immunity to the vicissitudes of disaster or conflict. Stress reactions are prominent in the immediate aftermath of a potentially traumatizing event, but these reactions typically decrease in the subsequent year and continue to attenuate even further after that. Resilience is common even in the individuals most highly exposed to the event and there are numerous protective buffers against emotional distress including perceived high levels of social support.

Three recent studies illustrate:

-          In a study of New York City residents soon after the September 11th attacks, many respondents did meet the various criteria for PTSD, especially those with high exposure to the events. However the majority of respondents – 65% – reported no trauma related symptoms.

-          A longitudinal study of Israelis assessed during and after rocket attacks focused on reported symptoms consistent with PTSD. In the period of attacks prevalence rates of PTSD were estimated at 20% but quickly declined to 3% two months after the attack.

-          Vacationing Swedish survivors of the 2004 tsunami were assessed at 14 month and 3 years after the disaster. While psychological morbidity scores are higher for those directly exposed or bereaved even 3 years after the disaster, all scores improve dramatically between the two periods.

These studies, and numerous others, also highlight the characteristics associated with resilience. Such factors include the presence of social support, marriage, higher education, and older age. Conversely, factors associated with persistent trauma reactions are gender (women consistently report more trauma reactivity), low education, previous diagnosis of a psychiatric illness, and further traumatic experiences after the event. (One challenge in the studies mentioned above is a high level of study attrition – roughly only 50% of contacted respondents agree to participate. Improving the response rate in future work will tell us whether these low participation rates lead to any bias in inference.)

The wider review by Bonanno and others focusing broadly on post-disaster impacts concludes with the interesting points:

1.       Disasters can cause psychological harm, but in a minority of exposed individuals. In the studies they review, the proportion exhibiting severe psychological distress such as PTSD rarely exceeds 30% of the exposed population.

2.       There are multiple outcome patterns in the population including those with 1 or 2 year delayed onset of trauma symptoms. However the most common reaction is only transient distress which doesn’t interfere with healthy functioning. This resilience is often exhibited in more than half of those exposed.

3.       Psychological impacts are mediated by a variety of risk and resilience factors such as those listed above.

4.       Typically, the resources of the affected community are mobilized after the disaster and survivors often receive immediate support from their relatives and friends. This is very important to note in the design of post-disaster aid. Communities should be maintained as much as possible and the use of community resources encouraged.

Given that resilience appears to be quite high in numerous populations exposed to a potentially traumatic event, as well as the fact that much trauma resolves naturally in the following year or two, certain policy implications appear quite clear. Blanket intervention in the immediate aftermath of a potentially traumatic may be, at best, mis-targeted and, at worse, distract from more critical aid efforts. Rather, some form of screening for at-risk individuals during the period of recovery, with the offer of effective individual or small-group interventions to those with likely PTSD, appears to be a much more effective strategy.


Jed Friedman

Senior Economist, Development Research Group, World Bank

Join the Conversation

March 14, 2012

While the examples you give do have some weight, I think it is quite dangerous to implement your suggetion of a 'screening' for at-risk individuals. The interesting and difficult problem surrounding PTSD is just that - it encompasses a host of various symptoms, many of which may be present but not vocalized by individuals because of societal expectations to 'just be okay', or in your words be 'resiliant'. Resiliance is associated with strength, while going for help is associated with weakness. This is what fundamentally needs to change. I work with individuals who are managing trauma, from post-war trauma, to sexual violence (which incidentally does occur in many disaster situations)to natural disasters. In my experience addressing PTSD from the outset leads to better recovery outcomes on a whole.

Emanuela Galasso
March 14, 2012

Jed, thanks for the interesting post.
Just wanted to flag a paper by Melissa Eccleston who is on the job market at Harvard Econ. She looks at in-utero exposure to maternal stress during 9/11 and finds significant negative effects on birth outcomes and early schooling outcomes. There might be long term consequences even if the maternal psychological bounces back to normal after the traumatic event.

Jennifer Lentfer
March 14, 2012

The lack of awareness about psychological recovery after disaster is my key issue with the video. Though I understand why it's gone viral, what does "the founder's inability to just 'stay' with Jacob in that dark, low moment" [when he talks about how much he misses his slain brother] say about the organization's approach? Read more here:

Jed Friedman
March 14, 2012

and the reminder that stigma around trauma and other forms of psychological distress presents a serious challenge for care. In no where in this post do I write that PTSD or other psycho-social disorder is not a serious problem. However many studies - those cited here and others - find that much of the trauma-related symptoms in the immediate aftermath of a post-traumatic event resolves itself naturally over a relatively short course of time. Of course this is not true for everyone and it is precisely these people that would most benefit from interventions and deserve to be targetted. The Bonanno review paper I cite has a very interesting discussion on effective interventions in disaster aftermath and according to the literature cited, blanket debriefing of exposed population doesn't appear to help very much. Hence the call for more targeted interventions at a later date that utilizes well-thought out and appropriate screening.

Jed Friedman
March 14, 2012

on what sounds like a very interesting paper!

Jed Friedman
March 14, 2012

... I sheepishly confess I haven't yet watched the Konyv video, but now am even more intrigued..

Alessandra Pigni
March 15, 2012

Many thanks for your much needed post! Glad to read this:

"Blanket intervention in the immediate aftermath of a potentially traumatic may be, at best, mis-targeted and, at worse, distract from more critical aid efforts". As Jennifer says, the lack of awareness is problematic. One way out is to seek tools that help all of us to cultivate that awareness, realising that at times we need to let ourselves and others go through the discomfort rather than anxiously trying to fix everything and everyone.

I'm a psychologist, former humanitarian professional and now staff-care consultant. I am in the process of publishing a White Paper Series on the psychological preparation of aid workers, and even for this group of people PTSD and trauma-related conditions are mistakenly considered "the issue". As human beings we are incredibly resilient! I'm investigating the topic of "post-traumatic growth" and I am integrating this concept in my vision for better humanitarian psychology for all, staff and so-called "beneficiaries". Here's a link to the White Paper Series

I will be happy to receive feedback and thoughts on this quite unique approach that I have to humanitarian psychology.

Sean Dalby
March 15, 2012

It's true that the accuracy of the results of these studies depends on a certain degree of self-awareness and courage within the members of the sample population to be honest about the psychological effects a disaster. We also probably see some of the social stigma against reporting trauma in the low turn out rate (~50%). (Of course, this is just a conjecture, albeit probable, too.) However, assessing whether or not someone is being honest in reporting their mental states also seems unnecessarily taxing for this kind of research, and so the assumption that people are accurate in their self-reports seems fair. We need to wary, as Jed notes, of this potentially biased sample, and leave the work of dismantling a pernicious social stigma to other projects and research.

That said, it seems safer to assume of the people that DID respond that they were accurate in their reports. This doesn't follow logically, strictly speaking, but it seems like a plausible story that people who have overcome social pressures to express their feelings would then report their mental status uninhibitedly (or at least they would be much more self-aware and honest than those refusing to participate). So, if this is true, then the longitudinal studies here DO suggest that recovery among this group is common and relatively quick. This is an encouraging result, and one that runs counter to many popular conceptions of the nature of psychological trauma caused by disasters.

Now, again as Jed says, none of this suggests that PTSD isn't serious. When it occurs it is indeed very serious and should be treated with expert medical care. The only suggestion is that *maybe* it doesn't occur as frequently as we think. We should not assume that everyone in a disaster-affected population is very likely to develop PTSD, for example, and perhaps reorient our intervention policies in light of this. The call for more research still rings out for sure and hopefully we will live in (or at least study) a culture that wholeheartedly approves of the expression of mental states during this future research.

March 28, 2012

To Whom It May Concern,

Pardon, but I only skimmed your post. I was wonder, in your research, did you come across any documentation or program to train the humanitarian aid or development workers who are sent to work in disaster/post-conflict areas?

Thank you for your time

Yours in development,
Jam tun (peace only)

Travis J. Warrington
Returned Peace Corps Volunteer/The Republic of The Gambia '08-'10
MA Degree Candidate in Sustainable International Development - 2012
MA Degree Candidate in Coexistence and Conflict - 2012
The Heller School for Social Policy and Management
Brandeis University
[email protected]